Re: v4l-dvb does not compile with kernel 2.6.34

2010-05-27 Thread Nima 
Last week I had the same problem. It gave me a headache trying to make
it work! So here is the instruction:

* First, follow the link below and download the .bz2 file.
http://mercurial.intuxication.org/hg/s2-liplianin

*Now, extract the files in the /usr/src/ directory and then, at the
prompt, type the following command:

wget 
http://www.forum.free-x.de/wbb/index.php?page=AttachmentattachmentID=479h=4f8e50d771ca4d3cd5d2fe38c7afe772c373313f;
-O SVT-SkyStarS2-driver-install.run.tar.bz2

*Extract the downloaded file and execute the .run file to patch the
v4l-dvb source code which you've downloaded by the first command.
./SVT-SkyStarS2-driver-install.run

Now, if it is patched correctly, you'll be able to run make and make
install. And that's it!

*You may encounter errors concerning a function named kzalloc. Let me
know if the compiling error occurred for you.

--
Yours sincerely,
Nima Mohammadi
--
To unsubscribe from this list: send the line unsubscribe linux-media in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: v4l-dvb does not compile with kernel 2.6.34

2010-05-26 Thread Helmut Auer 
Am 25.05.2010 23:59, schrieb Helmut Auer:
 Hello
 
 I just wanted to compile v4l-dvb for my Gen2VDR Ditribution with kernel 
 2.6.34, but it fails
 because many modules are missing:
 
 #include linux/slab.h
 
 and are getting errors like:
 
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c: In 
 function
 'free_firmware':
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c:252: 
 error: implicit
 declaration of function 'kfree'
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c: In 
 function
 'load_all_firmwares':
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c:314: 
 error: implicit
 declaration of function
 
 Am I missing something or is v4l-dvb broken ?
 
An easy patch for this problem is:

--- v4l/compat.h.org2010-05-26 22:22:31.0 +0200
+++ v4l/compat.h2010-05-26 22:22:43.0 +0200
@@ -28,6 +28,10 @@
 #include linux/i2c-dev.h
 #endif

+#if LINUX_VERSION_CODE = KERNEL_VERSION(2, 6, 33)
+#include linux/slab.h
+#endif
+
 #if LINUX_VERSION_CODE  KERNEL_VERSION(2, 6, 26)
 #ifdef CONFIG_PROC_FS
 #include linux/module.h


-- 
Helmut Auer, hel...@helmutauer.de
--
To unsubscribe from this list: send the line unsubscribe linux-media in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: v4l-dvb does not compile with kernel 2.6.34 [solved]

2010-05-26 Thread Helmut Auer 
Am 25.05.2010 23:59, schrieb Helmut Auer:
 Hello
 
 I just wanted to compile v4l-dvb for my Gen2VDR Ditribution with kernel 
 2.6.34, but it fails
 because many modules are missing:
 
 #include linux/slab.h
 
 and are getting errors like:
 
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c: In 
 function
 'free_firmware':
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c:252: 
 error: implicit
 declaration of function 'kfree'
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c: In 
 function
 'load_all_firmwares':
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c:314: 
 error: implicit
 declaration of function
 
 Am I missing something or is v4l-dvb broken ?
 
Here is the complete patch I am using:

--- v4l/compat.h.org2010-05-26 22:22:31.0 +0200
+++ v4l/compat.h2010-05-26 22:22:43.0 +0200
@@ -28,6 +28,10 @@
 #include linux/i2c-dev.h
 #endif

+#if LINUX_VERSION_CODE = KERNEL_VERSION(2, 6, 33)
+#include linux/slab.h
+#endif
+
 #if LINUX_VERSION_CODE  KERNEL_VERSION(2, 6, 26)
 #ifdef CONFIG_PROC_FS
 #include linux/module.h
--- linux/drivers/media/IR/ir-raw-event.c.org   2010-05-26 22:35:12.0 
+0200
+++ linux/drivers/media/IR/ir-raw-event.c   2010-05-26 22:35:43.0 
+0200
@@ -15,6 +15,7 @@
 #include linux/workqueue.h
 #include linux/spinlock.h
 #include linux/sched.h
+#include linux/slab.h
 #include ir-core-priv.h

 /* Define the max number of pulse/space transitions to buffer */
--- linux/drivers/media/dvb/dvb-core/dvb_frontend.h.org 2010-05-26 
23:06:50.0 +0200
+++ linux/drivers/media/dvb/dvb-core/dvb_frontend.h 2010-05-26 
23:06:21.0 +0200
@@ -39,6 +39,7 @@

 #include linux/dvb/frontend.h

+#include compat.h
 #include dvbdev.h

 struct dvb_frontend_tune_settings {
diff -upr /home/v4l/tmp/oldtree/drivers/media/radio/radio-tea5764.c
./drivers/media/radio/radio-tea5764.c
--- linux/drivers/media/radio/radio-tea5764.c.org   2010-05-25 
23:56:57.0 -0300
+++ linux/drivers/media/radio/radio-tea5764.c   2010-05-21 11:21:19.0 
-0300
@@ -32,6 +32,7 @@
  *  add RDS support
  */
 #include linux/kernel.h
+#include linux/slab.h
 #include linux/module.h
 #include linux/init.h/* Initdata 
*/
 #include linux/videodev2.h   /* kernel radio structs */

-- 
Helmut Auer, hel...@helmutauer.de
--
To unsubscribe from this list: send the line unsubscribe linux-media in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

v4l-dvb does not compile with kernel 2.6.34

2010-05-25 Thread Helmut Auer 
Hello

I just wanted to compile v4l-dvb for my Gen2VDR Ditribution with kernel 2.6.34, 
but it fails
because many modules are missing:

#include linux/slab.h

and are getting errors like:

/tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c: In 
function
'free_firmware':
/tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c:252: 
error: implicit
declaration of function 'kfree'
/tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c: In 
function
'load_all_firmwares':
/tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c:314: 
error: implicit
declaration of function

Am I missing something or is v4l-dvb broken ?

-- 
Helmut Auer, hel...@helmutauer.de
--
To unsubscribe from this list: send the line unsubscribe linux-media in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: v4l-dvb does not compile with kernel 2.6.34

2010-05-25 Thread VDR User 
On Tue, May 25, 2010 at 2:59 PM, Helmut Auer v...@helmutauer.de wrote:
 Hello

 I just wanted to compile v4l-dvb for my Gen2VDR Ditribution with kernel 
 2.6.34, but it fails
 because many modules are missing:

 #include linux/slab.h

 and are getting errors like:

 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c: In 
 function
 'free_firmware':
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c:252: 
 error: implicit
 declaration of function 'kfree'
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c: In 
 function
 'load_all_firmwares':
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c:314: 
 error: implicit
 declaration of function

 Am I missing something or is v4l-dvb broken ?

It's broken but at least you know the reason and how to manually fix it.
--
To unsubscribe from this list: send the line unsubscribe linux-media in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: v4l-dvb does not compile with kernel 2.6.34

2010-05-25 Thread hermann pitton 
Hi Helmut,

Am Dienstag, den 25.05.2010, 23:59 +0200 schrieb Helmut Auer:
 Hello
 
 I just wanted to compile v4l-dvb for my Gen2VDR Ditribution with kernel 
 2.6.34, but it fails
 because many modules are missing:
 
 #include linux/slab.h
 
 and are getting errors like:
 
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c: In 
 function
 'free_firmware':
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c:252: 
 error: implicit
 declaration of function 'kfree'
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c: In 
 function
 'load_all_firmwares':
 /tmp/portage/media-tv/v4l-dvb-hg-0.1-r3/work/v4l-dvb/v4l/tuner-xc2028.c:314: 
 error: implicit
 declaration of function
 
 Am I missing something or is v4l-dvb broken ?
 

I did not test on all, but had some rant few days ago and explanations
followed and I likely missed some RFCs in that direction previously.

If I do get it right now, the released vanilla 2.6.34 should be stable.

For the upcoming 2.6.35, like always, only the first upcoming rc1 is a
potential candidate for some first stability there.

On mercurial v4l-dvb, with all patch porting in this hybrid ;) situation
with mixed up- and backports, after the git branch is official, only
2.6.33 is considered to be stable for now.

There are still compatibility bugs on all earlier kernels I guess,
but /me agreed to have at least four weeks for fixing them.

An average kernel release cycle is about three months.

I asked, if we should not all go to latest stable vanilla and upcoming
rc candidates instead. By that mass of devices we have now, those with
relevant hardware must come back for testing on rc stuff anyway.

The decision for now was to keep backward compat, to allow to have more
testers, this was always my point of view over all the years too.
During all that, we changed a lot these days from hacked drivers to by
chip manufacturers and OEMs fully supported ones.

This is the root cause for going to git and includes to stay in sync
with other subsystems on git level doing the same.

For what is still trickling in from active users, I also have to say,
that this throws one back into time consuming efforts again, even more
time consuming than that times v4l and dvb did go out of sync on every
new kernel in the past and if it is only about to avoid latest vanilla
or upcoming rc stuff and keep staying on older kernels, is it worth it?

Testers will vote with their feet.

In other words, to keep bisecting reliable on Linus' git level is
already a huge task, to keep mercurial v4l-dvb bisecting with all the
compat stuff functional, ugh.

Also, no doubt, Hans' daily build reports are really helpful, but to
assume only seeing warnings there now, after a first fix for some
upper older kernel came in, and previously say eight did error out at
the same point, is no replacement for testing with real devices on every
such later kernel ...

The rest is pure luck and a rare case.

We likely should wait for two/three kernel release cycles, but I predict
Douglas will have a very hard job, if not much more people come in to
support backward compatibility on mercurial level.

Thanks for your report from the upper side ;)

Cheers,
Hermann








--
To unsubscribe from this list: send the line unsubscribe linux-media in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html